This conference presentation discusses a two-stage Zero-Shot Learning approach to classify a received signal originating from either a legitimate or unauthorized device in order to circumvent the threat posed by MAC-address spoofing attacks.
The authors present a paper that describes a two-stage Zero-Shot Learning (ZSL) approach to classify a received signal originating from either a legitimate or unauthorized device. This device is used to circumvent the threat of MAC-address spoofing, which is a common attack on wireless networks that allows adversaries to gain system access. Instead of relying on MAC addresses for admission control, fingerprinting allows devices to be classified before being granted access. With the ZSL approach, the classifier is first trained for classifying legitimate devices, where it learns discriminative features and the outlier detector uses those features to classify whether a new signature is an outlier. During the testing stage, an online clustering method is applied for grouping those identified unauthorized devices. The researchers’ approach allows 42% of unauthorized devices to be correctly identified and clustered.
Downloads
Related Datasets
Similar Publications
- Partners in Crisis: Improving Police Response to Individuals in Moments of Crisis by Providing Service Alternatives
- Activity Analysis in Crowded Environments Using Social Cues for Group Discovery and Human Interaction Modeling
- Validating a novel tool for coding body worn camera footage of police-community member interactions